This came up in a thread a couple of days ago about how you "hard limited" in SF. Early on in the piece someone suggested that hard limiting is achieved through using the "normalize" function in sound forge. My understanding is that hard limiting is through the dynamics functions. However, for the whole of the thread no one disputed that normalize was the right function. So I'm trying to confirm my sanity. Feel free to correct this.
My understanding is this:
"Normalize" amplifies the whole of the mix so that the loudest peak just hits the level that you're normalising to. Everything is amplified by the same amount (whether its + or -)
Hard Limiting is actually a (an extreme) form of compression. It leaves all signals the same except for those above a specified volume, which it limits (hard limits) at that volume.
So if you've got x db of dynamic range between loudest and softest sounds, you will still have that dynamic range after normalisation (unless you do some sort of extreme normalisation to -144 db or something
However, if you hard limit you will have less than x db of dynamic range because the troughs are unchanged but the peaks have been chopped off.
Sweet Dynamic Range